When information is stored in any sort of structure such as a database, if the information is ever transferred to another structure, there are considerations that must be taken into account such that all of the information is retained and is still accessible. If the first structure and the second structure are exactly the same with the same field names, then there should not be any issues with copying since the data will be accessible as it was before. For example, a database stores contact information, and the database fields include business fax, emergency contacts and home phone. If a company upgrades to a new system that has a new database but the new database still has the same fields of business fax, emergency contacts and home phone, the transition should be very smooth since the data will be located in the same fields.
However, when the databases are not the same, specifically the fields are not exactly the same, a methodology must be used to ensure the data is not lost and is easily accessible. One methodology is referred to as fixed field mapping where fields are fixedly mapped from one structure to another structure. For example, if the content of the employee database as described above is moved to a new database that has the fields of fax, home phone and other, a fixed mapping could include mapping business fax to fax, home phone to home phone and emergency contacts to other. While this adequately solves the problem of ensuring all of the data is transferred, it may lead to data being placed in a location that a user does not like. Furthermore, fixed field mapping requires either prior knowledge of the fields in the second database or assumptions of what fields are in the second database. Prior knowledge of the fields requires additional time and effort to make the transfer of data, and assumptions have the potential of being highly inaccurate.